ResearchPapers/fawkes
ResearchPapers/fawkes
2
0
Fork 0
mirror of https://github.com/Shawn-Shan/fawkes.git synced 2024-12-22 07:09:33 +05:30
Code

0.3

Browse Source
This commit is contained in:
Shawn-Shan 2020-07-29 10:31:34 -05:00
parent e25cff66ac
commit 1a810c8fb7
2 changed files with 2 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
fawkes/protection.py
View File

@ -80,7 +80,7 @@ class Fawkes(object):
max_step = 1000 max_step = 1000
lr = 8 lr = 8
else: else:
raise Exception("mode must be one of 'low', 'mid', 'high', 'ultra', 'custom'") raise Exception("mode must be one of 'min', 'low', 'mid', 'high', 'ultra', 'custom'")
return th, max_step, lr return th, max_step, lr
def run_protection(self, image_paths, mode='min', th=0.04, sd=1e9, lr=10, max_step=500, batch_size=1, format='png', def run_protection(self, image_paths, mode='min', th=0.04, sd=1e9, lr=10, max_step=500, batch_size=1, format='png',

9
fawkes/utils.py
View File

@ -233,7 +233,6 @@ def load_victim_model(number_classes, teacher_model=None, end2end=False):
def resize(img, sz): def resize(img, sz):
assert np.min(img) >= 0 and np.max(img) <= 255.0 assert np.min(img) >= 0 and np.max(img) <= 255.0
from keras.preprocessing import image from keras.preprocessing import image
im_data = image.array_to_img(img).resize((sz[1], sz[0])) im_data = image.array_to_img(img).resize((sz[1], sz[0]))
im_data = image.img_to_array(im_data) im_data = image.img_to_array(im_data)
@ -431,18 +430,12 @@ def dump_image(x, filename, format="png", scale=False):
def load_embeddings(feature_extractors_names): def load_embeddings(feature_extractors_names):
model_dir = os.path.join(os.path.expanduser('~'), '.fawkes') model_dir = os.path.join(os.path.expanduser('~'), '.fawkes')
dictionaries = []
for extractor_name in feature_extractors_names: for extractor_name in feature_extractors_names:
fp = gzip.open(os.path.join(model_dir, "{}_emb.p.gz".format(extractor_name)), 'rb') fp = gzip.open(os.path.join(model_dir, "{}_emb.p.gz".format(extractor_name)), 'rb')
path2emb = pickle.load(fp) path2emb = pickle.load(fp)
fp.close() fp.close()
dictionaries.append(path2emb)
merge_dict = {} return path2emb
for k in dictionaries[0].keys():
cur_emb = [dic[k] for dic in dictionaries]
merge_dict[k] = np.concatenate(cur_emb)
return merge_dict
def extractor_ls_predict(feature_extractors_ls, X): def extractor_ls_predict(feature_extractors_ls, X):